U.S. Pat. No. 3,597,051 to Copeland, incorporated herein in its entirety by this reference, illustrates an exemplary streak retinoscope assembly. One such commercial assembly, the "Optec 360" retinoscope 10 of FIG. 1, includes a thumb slide 12 adapted to move relative to the retinoscope shaft. As shown in FIG. 2A, advancing slide 12 to its upper position causes the light rays emanating from the retinoscope to be approximately parallel. FIG. 2B, by contrast, illustrates the convergence of the light rays that occurs when slide 12 is moved to its lower position.
Current refraction techniques use the streak retinoscope for neutralizing optical errors. Such techniques are described in, for example, Videotape No. 5063 of the American Academy of Ophthalmology's Continuing Ophthalmic Video Education series, entitled "Retinoscopy: Plus Cylinder Technique," and require use of a phoropter, trial frames, or additional lenses. According to these techniques, the slide of the retinoscope remains in the upper position throughout the retinoscopic process. Only if greater than one diopter of astigmatic error is present do these techniques suggest "enhancing" the patient's cylinder power by lowering the retinoscope slide. By contrast, "enhancing" the streak to estimate the sphere power does not occur.
After the practitioner neutralizes the patient's error by changing the phoropter or trial frame lenses, the values of those lenses are consulted to determine the patient's ocular correction. These refraction techniques essentially use the phoropter or trial frame lenses to make the light rays emanating from the retinoscope conjugate to the patient's fundus. (Although these rays are generally assumed to be parallel for these techniques, my recent empirical studies suggest that they are not for commercially-available retinoscopes.) Doing so in turn causes the rays backscattered by the patient's fundus to be conjugate to the practitioner's eye, permitting neutralization of optical error at a specified working distance.
According to Videotape No. 5063, prior retinoscopic estimating techniques were complex and thus rarely learned or used by the average practitioner. Such "two-handed" techniques require the practitioner to rotate a collar or sleeve on the retinoscope while simultaneously moving the slide up and down, effectively creating "spiral" movement of the slide and sleeve. To perform these techniques, moreover, the practitioner must move back and forth relative to the patient, thereby alternately approaching and receding from the eye under examination. The patient's optical error is then typically estimated based on the width of the focused streak of light emanating from the retinoscope as seen by the practitioner on the patient's pupil. Following "straddling" and other movements of the streak, the patient's cylindrical error axis can ultimately be estimated by comparing the longitudinal axis of the streak to a scale on the phoropter. Moreover, as discussed on page 21 of Dr. J. C. Copeland's manual for "Steak Retinoscopy" printed by Optec, Inc. (which manual is incorporated herein in its entirety by this reference), these prior estimating techniques were "best . . . perform[ed] . . . on the naked eye" and thus did not involve overrefraction of existing prescription lenses.